AI Briefing#generative-models
The latest AI and developer news about #generative-models, with the original source and a short summary.
The latest AI and developer news about #generative-models, with the original source and a short summary.
A model called VICIS has been proposed that enables Vision-Language Models to infer visual concepts from example images and apply them.
Introduces the latest trends in generative models, including the Score-based models and VAEBM that drew attention at ICLR 2021.
This introduces generative models and SDE-based Score-based Generative Modeling, key trends at ICLR 2021.
This paper introduces a new methodology that leverages the characteristics of Hierarchical VAE to improve Out-of-distribution detection performance.
This piece analyzes the principles of Diffusion Models, a core component of generative models, and the differences among various generative models.
This review focuses on the Diffusion Model, one of the most-noticed generative models among the diverse works presented at NeurIPS 2021, and covers the key research on it.
LG AI Research is researching a loop-based music generation model to help with motif generation, which is at the core of music composition.
LG AI Research proposed Projection Regret, a technique that uses a Diffusion model to reduce background noise and improve anomaly detection performance.
DiScoFormer was released, estimating a data's Density and Score in a single computation without retraining.
Qwen-Image-Flash demonstrates the importance of optimizing data composition and training pipelines, beyond objective functions, for accelerating image generation models.
Proposed a new Flow Matching method that steers the generative flow using a reference set.
It explains how flow maps directly predict the integral of diffusion sampling.
STARFlow-V demonstrates high-quality autoregressive video generation without diffusion.
Microsoft unveiled **TRELLIS.2**, supporting high-resolution image-to-3D generation.
Implemented natural likeness using an HG-DPO-based backbone model and the Instantbooth module, validated via a photo booth at if(kakaoAI)2024.
To overcome the slow generation speed of diffusion models, we propose **Consistency Models**, which map noise directly to data.
By introducing a method that learns directly from data without distillation, the generation quality of the Consistency Model has been significantly improved.